QuantumATK Forum
QuantumATK => General Questions and Answers => Topic started by: renren123123 on December 29, 2008, 04:12
-
I have calculated a model of two different electrode system, and I have obtained some results such as current ,transimission and DOS, however, I am going to get more results such as MPSH,LDOS, charge density and so on, I face to a large problem: originally 2 different electrode unit cell with different coordinates especially along z direction will overlap in VNL due to its lattice consist, how to deal with this?
thank you very much!!!
-
First of all, it is a good idea to the notes on heterogeneous twoprobe geometry
which can be found in the online manual here (http://quantumwise.com/documents/manuals/ATK-2008.10/ref.twoprobeconfiguration.html#ref.twoprobeconfiguration.notesl).
Scroll down to the Notes section and read this section and if you still have questions after reading this part.
Then I can perhaps take a look at your problem with you :)
P.S: Despair not, it is only the first couple of times It is tricky to setup a heterogeneous twoprobe, and there should not be any problem with your systems :)
Edit:
Updated link.
-
Thank you for your advice! however, I still face to a problem:how to define right electrode unit cell
for example [[ 30.0000000, 0. , 0. ],
[ 0. , 30.0000000, 0. ],
[ 0. , 0. , 5.06373621 ]]*Angstrom
we only can define it from origin point 0, it will display: left and right electrode unit cell will overlap in
VNL!!! How to define right electrode unit cell not from origin point 0?
-
ATK aligns the first atom of scattering region and last atom of electrode automatically. What do you mean by saying " left and right electrode overlap in vnl".
-
Hey renren123123.
As Hasan Sahin says.
You simply define the two electrodes as they were alone in the world. Then when you create your twoprobe you will set those equivalent_atoms,
and then ATK will align the geometry into a complete twoprobe, and the electrodes will not overlap in VNL, but be aligned according to the equivalent atoms as described in the manual section.
If you want, I can try to make a minature example for you of this, so you can see it.
-
can you give me an example of this? thank you very much! my qq:823649089 email: [email protected]
my script partly as following:
however, I see its configuration in VNL, left and right electrode unit cell will overlap as shown by frame
-
Well both the electrodes are correctly setup, however what is properly not correct is
equivalent_atoms=([-1,31],[-1,-1])
But I can't be sure until share some of the scattering elements since this can not be visualized. I understand you want to protect the
the exact device you are trying to model, but if we are to be able to help you, you need to provide the outer most egde of the scattering region.
If you will like to try to debug this yourself, then check your equivalent atoms:
- The atom of the scattering region with index 31 ( the 32th atom ) must be the same as the the last element of the electrode, and translated by the left electrode lattices constant along the z-axises
- The last element of the scattering region must also be the last element of the right electrode translated by the right electrode lattice constant along the z-axises.
It means that the atom with index 31 ( the 32th ) in the scattering region must the same as the last
-
For your convience, and since it is chrismas and I have too much time in these days - I have given a shot at
trying to make a heterogeneous twoprobe that could be of interest to you :)
-
let me seeļ¼Thanks a lot! can you give me your email or other affiliation ?
-
If you click my profile name, I have made my email public.